Cathepsin H increases the risk of diabetic retinopathy: evidence from Mendelian randomization and bioinformatic analysis

Abstract Background While lysosomal cathepsins are crucial in cellular homeostasis and may contribute to diabetic chronic complications, their precise causal relationships remain insufficiently characterized. Methods This study employed comprehensive multidimensional Mendelian randomization (MR) analyses to investigate potential causal associations between nine cathepsins (B, E, F, G, H, L2, O, S, and Z) and six diabetic chronic complications encompassing both microvascular (nephropathy, retinopathy, proliferative retinopathy, maculopathy, neuropathy) and macrovascular manifestations (peripheral angiopathy). The analytical framework incorporated generalized summary-data-based MR (GSMR), univariable MR, multivariable MR, summary-data-based MR (SMR), and cis-expression quantitative trait locus (cis-eQTL) MR approaches, utilizing data derived from publicly available genome-wide association studies (GWAS). To further characterize the biological relevance of identified cathepsins, we conducted multi-omics investigations including gene set enrichment analysis, CIBERSORT, single-cell RNA sequencing analysis to explore the expression level, immune infiltration and biological function of identified cathepsins. Furthermore, we performed mediation analyses to assess whether immune cells potentially mediate the causal pathways linking identified cathepsins to diabetic complications. Results Both GSMR and univariable MR showed that Cathepsin H levels were causally associated with increased risks of overall diabetic retinopathy (DR), proliferative diabetic retinopathy (PDR), and diabetic maculopathy, validated by SMR and cis-eQTL MR analyses. Multivariable MR analysis further confirmed the independent causal role of Cathepsin H in PDR {inverse variance weighted (IVW): p = 0.003, OR = 1.054, 95% CI = 1.017–1.093} and diabetic maculopathy (IVW: p = 0.022, OR = 1.068, 95% CI = 1.009–1.130). Sensitivity analyses indicated no significant heterogeneity or pleiotropy for any of these cause associations. Bioinformatic analyses revealed upregulated Cathepsin H expression in DR patients, enriched in immune-inflammatory pathways. Single-cell RNA sequencing further highlighted its specific overexpression in NK cells of DR. No mediation by immunophenotypes was observed. Conclusions Our integrative approach establishes Cathepsin H as a causal driver for DR and its subtypes, highlighting its potential as a therapeutic target and biomarker.